Abstract

Objective—To develop a practitioner led definition of a prescribing error for use in quantitative studies of their incidence.

Design—Two stage Delphi technique.

Subjects—A panel of 34 UK judges, which included physicians, surgeons, pharmacists, nurses and risk managers.

Main outcome measures—The extent to which judges agreed with a general definition of a prescribing error, and the extent to which they agreed that each of 42 scenarios represented a prescribing error.

Results—Responses were obtained from 30 (88%) of 34 judges in the first Delphi round, and from 26 (87%) of 30 in the second round. The general definition of a prescribing error was accepted. The panel reached consensus that 24 of the 42 scenarios should be included as prescribing errors and that five should be excluded. In general, transcription errors, failure to communicate essential information, and the use of drugs or doses inappropriate for the individual patient were considered prescribing errors; deviations from policies or guidelines were not.

Conclusions—Health care professionals are in broad agreement about the types of events that should be included and excluded as prescribing errors. A general definition of a prescribing error has been developed, together with more detailed guidance regarding the types of events that should be included. This definition allows the comparison of prescribing error rates among different prescribing systems and different hospitals, and is suitable for use in both research and clinical governance initiatives.

(Quality in Health Care 2000;9:232–237)